Post on 12-Apr-2020
Instructions for use
Title Study of Deuterium Isotope Separation by PEFC
Author(s) Shibuya, Shota; Matsushima, Hisayoshi; Ueda, Mikito
Citation Journal of the electrochemical society, 163(7), F704-F707https://doi.org/10.1149/2.1321607jes
Issue Date 2016-07-20
Doc URL http://hdl.handle.net/2115/62424
Rights© The Electrochemical Society, Inc. 2016. All rights reserved. Except as provided under U.S. copyright law, this workmay not be reproduced, resold, distributed, or modified without the express permission of The Electrochemical Society(ECS). The archival version of this work was published in Journal of the electrochemical society, volume 163, Issue7, pp. F704-F707, 2016.
Type article (author version)
File Information Manuscript-ECS-revised.pdf
Hokkaido University Collection of Scholarly and Academic Papers : HUSCAP
1
Study of Deuterium Isotope Separation by PEFC Shota Shibuya, Hisayoshi Matsushima* and Mikito Ueda
1Faculty of Engineering, Hokkaido University,
Kita 13 Nishi 8, Sapporo, Hokkaido 060-8628, Japan *Corresponding Author: matsushima@eng.hokudai.ac.jp
Tel. & Fax +81-11-7066352
2
ABSTRACT
Hydrogen evolution and oxidation reactions were studied on a polycrystalline platinum
electrode in 0.05 M D2SO4 solution at 298 K by using a rotating disk electrode,
focusing on the kinetic isotope effect on reactivity. The polarization measurements in
the evolution reaction regime, on the one hand, revealed two Tafel regions: at low
overpotentials close to the equilibrium potential, the Tafel slope was 0.039 V dec-1,
suggesting a Volmer-Tafel mechanism; and at potentials around 0.05 V or lower, a slope
of 0.168 V dec-1 indicated that a transition to a Heyrovsky-Tafel mechanism occurred.
In the oxidation reaction regime, on the other hand, a single Tafel slope of 0.055 V dec-1
was observed at potentials of 0.08 V or lower. Comparing between the deuterium and
protium data for the kinetic factors supported the presence of an isotope effect, whereby
the deuterium was more easily dissociated on the platinum electrode. The present
kinetic result supported the deuterium separation of polymer electrolyte fuel cell
(PEFC) in which the mixture gases of H2 and D2 were purged. The deuterium separation
factor increased with increasing in the hydrogen utilization. This suggests a new isotope
separation technique and also provides useful data for fuel cells.
Keywords: isotope effect; separation factor; deuterium; electrolysis; hydrogen oxide
reaction; hydrogen evolution reaction; fuel cell
3
1. Introduction Hydrogen isotopes are important for fission and fusion reactors [1].
Deuterium, which comprises 0.0149% of hydrogen, is found in natural water, whereas
tritium is artificially produced as a by-product of fission reactors.
Techniques of hydrogen isotope separation include several approaches such as
distillation [2], chemical exchange [3], quantum sieving [4], and so on [5-7]. Whiticar
reported that the bacterial reduction of hydride is associated with a kinetic isotope effect
resulting in the enrichment of deuterium [8]. However, most methods are ineffective
and have low separation factors. Large-scale plants must deal with enormous volumes
of water for isotope enrichment, and especially a new separation technology for tritium
at the Fukushima Daiichi Nuclear Power Plant in Japan is urgently required.
One of the most practical methods is CECE (combined electrolysis catalytic
exchange) method [9-13]. This method combines electrolysis with chemical exchange
reaction. Hydrogen generated in electrolysis is passed into a column mixed with kogel
catalyst and Dixon rings. The isotope exchange reaction occurs at the catalytic site,
thereby deuterium or tritium can be enriched. This is one of the practical methods due to
the reuse of generated hydrogen and high separation factor. However, the electrolysis
consumes a huge amount of electric energy and the catalyst requires a lot of noble
metal.
Water electrolysis has long been studied for deuterium separation [14-20].
This method has reported the highest separation factor. For example, separation factors
of 5–8 are reported for platinum [21] and the highest values of 10–13 are for iron metal
[22]. The separation mechanism is attributed to the slower discharge reaction of
deuterons than protons. As well as CECE method, the electrolysis is not suitable for
large production of hydrogen isotopes. Therefore, we have been proposing a new
electrolytic deuterium separation system using a fuel cell. It is suggested by calculation
that the new system remarkably should reduce electric energy consumption for heavy
water production by reusing the hydrogen and oxygen gases at a fuel cell [23]. However,
the hydrogen isotope effect of a fuel cell is not experimentally investigated very much.
Recently Oi et al. reported that the hydrogen isotope could occur at a fuel cell [24, 25].
The deuterium separation factor was simulated to be from 3.5 to 4.0 depending on the
hydrogen utilization.
The kinetics of the hydrogen evolution reaction (HER) is still an important
4
topic in electrochemistry for both practical applications and fundamental studies. Thus,
there are many studies of the kinetic factors for polycrystalline platinum electrodes
[26-28]. However, there are a few experimental studies on the kinetic factors of the
deuterium evolution reaction (DER) [29]. Previously, Bockris et al. reported that the
Tafel slope of DER was 0.026 V dec-1, which was similar to that of HER (0.029 V dec-1),
whereas the exchange current of DER was much smaller than that of HER [30].
The hydrogen oxidation reaction (HOR), which is the reverse reaction of HER
and the main reaction in a fuel cell, remains the subject of ongoing investigation [31-34].
Moreover, the fundamental kinetic data of the deuterium oxidation reaction (DOR) have
not been measured. In this study, we investigate the isotope effect in the hydrogen
electrode reaction on a platinum electrode under high-concentration deuterium
conditions and then demonstrate the deuterium separation by using PEFC.
2. Experimental 2.1 Measurement of hydrogen electrode reactions
The measurements were conducted under a pure hydrogen or deuterium gas
atmosphere. The working electrode was a 0.785 cm2 polycrystalline platinum disk (Pt
99.99%, Nilaco Corp.) and the counter electrode was a coiled platinum wire. The
electrodes were cleaned by ultrasonication in acetone for 30 min and washing in
distilled water. They were annealed for 5 min under hydrogen atmosphere and quenched
in distilled water in order to activate the surface. The reference electrode was
Hg/Hg2SO4 saturated with potassium sulfate. A rotating disk electrode (Model 636, EG
& G) was used with rotation rates of 900, 1600, 2500, and 3600 rpm [35, 36]. The
polarization curves were recorded under potential control by a potentiostat (Model
SP-50, Biologic). A sulfuric acid solution of 0.05 M (98 wt% H2SO4, Wako Pure
Chemical Ind.) was adjusted with distilled water for the protium experiment. A
deuterated sulfuric acid solution of 0.05 M (98 wt% D2SO4, Sigma-Aldrich) was made
with heavy water (99.9 at% D2O, Isotech) for the deuterium experiment. Pure hydrogen
or deuterium gas was bubbled through the electrolyte for 2 h. The solution temperature
was maintained at 298 K.
2.2 Measurement of the separation factor at PEFC
JARI standard cell (FC Development Corp.) was employed as PEFC. A
5
membrane electrode assembly (50×50 mm) was composed of Nafion electrolyte
(NRE-212) and two catalytic layers loaded with platinum catalyst (Pt 0.52 mg cm-2).
PEFC was operated at 298 K and the power generation was controlled under constant
current mode (0.0 A - 1.2 A) adjusted by a variable resistor unit (Kikusui Electronics
Corp., PLZ 164WA).
Deuterium separation factor, α, of PEFC was measured by quadrupole mass spectrometer (QMS) (Ulvac Corp., Qulee-HGM 202). Figure 1 shows the schematic
experimental setup for obtaining α at PEFC. The humidified O2 gas of 50 mL min-1 was supplied into cathode. The mixture gases of H2 and D2 were used at anode. The mixing
ratio was adjusted by mass flow controller (Alicat Corp., MC-200SCCM-D). The flow
rate of H2 and D2 was 20 mL min-1 and 2 mL min-1, respectively. QMS monitored the
gas component exhausting from anode before and after PEFC operation. It recorded the
ion current of mass number (m = 2, 3 and 4) for 8 hours at which the ion current reached
steady state value.
3. Results and Discussion 3.1 Kinetics of Hydrogen Electrode Reactions
The fundamental experiment of hydrogen electrode reaction was conducted
by using rotating disk electrode under a flow of hydrogen or deuterium gas to exclude
oxygen. Before the electrochemical measurements, the open circuit potential (OCP) was
recorded as -0.725 V in H2SO4 and -0.730 V in D2SO4 electrolyte. The small potential
difference of 0.005 V is explained by the Gibbs free energy of deuterium corresponding
to -0.013 V at 298 K [29].
The polarization curves were measured by linear sweep voltammetry at a scan
rate of ±0.01 V s-1. Figure 2 shows the current density for HER/HOR and DER/DOR for
a scanning range of ±0.3 V from OCP. The reduction reaction currents were independent
of the rotation rate in both electrolytes, whereas the oxidation currents were
significantly affected by forced convection at more than -0.65 and -0.70 V HER/HOR
and DER/DOR, respectively. The limiting current densities, which were induced by the
diffusional limitation of the dissolved hydrogen gas, appeared at high electrode
potentials (E > -0.5 V).
3.1.1 HER and DER
6
The Tafel slopes for HER and DER were determined at low overpotentials (0
V ≥ η ≥ -0.15 V), as seen in Fig. 3. Two Tafel slopes were observed for HER in H2SO4
solution. The slope changed around η = -0.05 V. The value was 0.034 V dec-1 at low
overpotentials and 0.143 V dec-1 at high overpotentials. The mechanism of HER on
polycrystalline platinum has been thoroughly investigated [34-37]. The low slope value
corresponds to the Volmer-Tafel mechanism and the high slope value corresponds to the
Heyrovsky-Tafel mechanism. The Tafel slopes of DER, too, could be divided into two
Tafel regions. The slope value was 0.039 V dec-1 at low overpotentials (η ≥ -0.05 V) and
0.168 V dec-1 at high overpotentials. The slope values of HER and DER under
Volmer-Tafel conditions were similar, whereas the difference between the slopes
increased under the Heyrovsky-Tafel regime in this work.
The Tafel slopes were extrapolated to the x-axis to obtain the exchange
current density. The values for HER, i0(H+), and DER, i0(D+), were 0.031 and 0.018 mA
cm-2, respectively. The ratio of i0(H+)/i0(D+) was 1.72, which was in good agreement with
other result [30]. The present values of the exchange current density are larger than
general values [28-30]. This might be explained by surface roughness. Thus, we simply
calculated the surface area as the geometry area.
3.1.2 HOR and DOR
The anodic currents did not depend on the rotational rate at η ≤ 0.08 V, which
suggests that the oxidation reactions of hydrogen or deuterium were kinetically
controlled. As the anodic potential increased, the reactions were under a mixture of
kinetic and diffusion control. To clarify the Tafel region, the current density was
corrected by the limiting current [34]. Fig. 4 shows Tafel plots of mass
transfer-corrected currents for the HOR and DOR at low anodic overpotentials.
There was one Tafel region at low overpotentials (η ≤ 0.08 V) for HOR and
DOR in contrast to HER. The Tafel region of HOR began at η ≈ 0.04 V. The slope value
of HOR was 0.091 V dec-1, which was much larger than that of the Volmer-Tafel
mechanism. This might be explained by the slow dissociation of hydrogen gas
molecules, although more detailed studies are required to confirm this. The Tafel slope
of DOR was observed from a lower overpotential (η ≥ 0.03 V). The slope value was
0.055 V dec-1, which was significantly smaller than that of HOR. Although the
exchange current densities in Fig. 4 did not match those in Fig. 3, the ratio of i0(H2)/i0(D2)
7
was 0.46.
Comparison between the deuterium and protium data for the kinetic factors
indicated an isotope effect: deuterium was more easily dissociated on a platinum
electrode, suggesting the deuterium separation at a fuel cell. Therefore, we
demonstrated the deuterium separation by using PEFC at the next section.
3.2 Deuterium Separation by PEFC
The mixture gases was directly connected to QMS (gas line I in Fig. 1). The
ratio of H and D was checked. The results of QMS showed that the gas composition was
very stable during the measurement. As seen in Fig. 5 (a), the ion current of mass
number m = 4 was about 1/10 times less than that of m = 2.
Once the mixture gas was introduced to anode at PEFC (gas line II in Fig. 2),
the ratio of m = 3 was significantly increased instead of decreasing in that of m = 4 (Fig.
5 (b)). This reversal phenomenon of ion current between m = 3 and m = 4 was unrelated
to the power generation. Taking into account of the unchanged atomic ratio of H and D
might give us one explanation. When the both gases of H2 and D2 are passed to the
catalytic layer, HD gas (m = 3) is generated by isotope exchange reaction as expressed
in Eq. (1).
H2 + D2 → 2HD (1)
The exchange reaction is reported to occur on platinum surface [38, 39]. The almost all
molecules of H2 and D2 introduced into PEFC were converted to HD since the ion
current of m = 4 was remarkably dropped. The gas diffusion layer and nano-sized Pt
catalyst, which are researched and developed for PEFC, might contribute to the high
activity for Eq. (1).
The ratio of gas components changed after PEFC was operated at 1.2 A
(output voltage = 0.78 V) as seen in Fig. 5 (b). The ion current of HD and D2 decreased
while H2 one slightly increased. It took 1.5 hours to reach steady state. The transient
time might be attributed to the replacement of QMS chamber by the reacted gases. The
gases containing deuterium was preferentially consumed at PEFC. It was found that the
composition ratio of D2 decreased larger than that of HD. This coincided with the
kinetic isotope effect of which the degree is enlarged with increasing in mass number
8
[40].
The deuterium separation factor was obtained by measuring the outlet gas
(gas lines II in Fig. 1) from PEFC. The factor was calculated by the following equation,
α = ([H]/[D])a / ([H]/[D])b (2)
where [H] and [D] are atomic fractions of protium and deuterium, the subscript (a) and
(b) refers to after and before starting the power generation.
The separation factor was calculated at 5 hours after starting PEFC. Figure 6
shows α depending on the output current. The factor value increased from α = 1.1 at
0.05 A to α = 4.6 at 1.2 A. The isotope condensation toward the produced water was accelerated with increasing in the output current. This is probably attributed to the
hydrogen utilization. Thus, low fuel utility rises the unreacted mixture gases which can
pass only through the gas diffusion layer. Although the factor of 4.6 is higher than other
deuterium separation methods such as CECE, the good separation performance is
expectedly achieved by developing catalyst and operation conditions, since the present
fuel utilization of 40 % was still low.
4. Conclusions In order to investigate possibility of the hydrogen isotope separation by PEFC,
the kinetic electrochemical factors were investigated and the deuterium separation was
demonstrated practically.
The exchange current density and Tafel slope measured by a rotating disk
electrodes indicated the isotope effect on the hydrogen oxidation reaction as well as on
the evolution reaction. The values of exchange current density and Tafel slope in HOR
were 0.130 mA cm-2 and 0.091 V dec-1, while the values in DOR were 0.057 mA cm-2
and 0.055 V dec-1, respectively.
PEFC with Nafion MEA was galvanostatically operated under introducing the
mixture gases of H2 and D2. The deuterium separation factor significantly increased
with increasing in the output current. The maximum value was 4.6 at 1.2 A (0.93 W at
298 K) at the present experiment. Corresponding to the kinetic results, the deuterium
could be separated as the produced water through the oxidation reaction. Compared
with other separation processes, the large value of α might be attributed to the mixture
9
effects of the electrochemical reactions and membrane. It will be clarified by
investigating the isotope ratio of the produced water the in the future.
ACKNOWLEDGMENT
The authors appreciate financial support from the Ministry of Education,
Culture, Sports, Science and Technology (Project No. 26870234). One of the authors, H.
Matsushima, wishes to express his sincere gratitude to Inamori Foundation in Japan.
10
REFERENCE
1. S. J. Zinkle, L. L. Snead, Ann. Rev. Mater. Res., 44, 241 (2014).
2. Y. Iwai, H. Yoshida, T. Yamanishi, S. Senrui, M. Nishi, Fusion Eng. Des., 49, 847
(2000).
3. F. Huang, C. G. Meng, Int. J. Hydrog. Energy, 35, 6108 (2010)
4. S. A. FitzGerald, C. J. Pierce, J. L. C. Rowsell, E. D. Bloch, J. A. Mason, J. Am.
Chem. Soc., 135, 9458 (2013).
5. Y. Sakharovski, V. Tkachenko, J. Radioanal. Nucl. Chem., 304, 357 (2015).
6. M. Hankel, Y. Jiao, A. Du, S. K. Gray, S. C. Smith, J. Phys. Chem., 116, 6672
(2012).
7. K. Watanabe, M. Matsuyama, T. Kobayashi, S. Taguchi, J. Alloy Com., 257, 278
(1997).
8. M. J. Whiticar, Chem. Geol., 161, 291 (1999).
9. I. A. Alekseev, S. D. Bondarenko, O. A. Fedorchenko et al., Fusion Sci. Technol.,
60, 1117 (2011)
10. Y. Sun, H. Y. Wang, G. Sang, Fusion Eng. Des., 83, 1400 (2008).
11. T. Sugiyama, Y. Asakura, T. Uda, T. Shiozaki, Y. Enokida, I. Yamamoto, Fusion
Eng. Des., 81, 833 (2006).
12. J. M. Miller, S. L. Celovsky, A. E. Everatt, W. R. C. Graham, J. R. R. Tremblay, ,
Fusion Sci. Technol., 41, 1077 (2002).
13. H. K. Rae, Trans. ANS, 23, 65 (1976).
14. S. S. Mijanic, A. D. Maksic, N. I. Potkonjak, M. P. M. Kaninski, Electrochem.
Commun., 9, 2179 (2007).
15. B. Dandapan, M. Fleischm, J. Electroanal. Chem., 39, 315(2002).
16. L. I. Krishtalik, Electrochim. Acta, 46, 2949 (2001).
17. D. L. J. Stojic, S. S. Miljanic, T. D. Grozdic, L. T. Petkovska, M. M. Jaksic, Int. J.
Hydrog. Energy, 25, 819 (2000).
18. J. Sperling, H. Tributsch, R. M. F. Streffer, T. A. Redah, C. A. C. Dreismann, J.
Electroanal. Chem., 477, 62 (1999).
19. T. Green, D. Britz, J. Electroanal. Chem., 412, 59 (1996).
20. B. Topley, H. Eyring, J. Chem. Physics, 2, 17494 (1934).
21. M. Hammerli, J. P. Mislan, W. J. Olmstead, J. Electrochem. Soc., 117, 751 (1970).
22. H. Matsushima, T. Nohira, Y. Ito, Electrochim. Acta, 49, 4181 (2004).
11
23. H. Matsushima, T. Nohira, T. Kitabata, Y. Ito, Energy, 30, 2413 (2005).
24. S. Yanase, T. Oi., Z. Naturforsch, 70, 429 (2015).
25. S. Yanase, T. Oi, J. Nucl. Sci. Technol., 50, 808 (2013).
26. P. M. Quaino, M. R. Chialvo, A. C. Chialvo, Electrochim. Acta, 52, 7396 (2007).
27. M. C. Tavares, S. A. S. Machado, L. H. Mazo, Electrochim. Acta, 46, 4359 (2001).
28. H. Kita, J. Electrochem. Soc., 113, 1095 (1966).
29. J. D. E. McIntyre, M. Salomon, J. Phys. Chem., 72, 2431 (1968).
30. J. O’M. Bockris, D. F. A. Koch, J. Phys. Chem., 65, 1941 (1961).
31. T. Shinagawa, A. T. Garcia-Esparza, K. Takanabe, ChemElectroChem, 1, 1497,
(2014).
32. J. Durst, A. Siebel, C. Simon, F. Hasche, J. Herranz, H. A. Gasteiger, Energy
Environ. Sci., 7, 2255 (2014).
33. W. Sheng, H. A. Gasteiger, Y. S. Horn, J. Electrochem. Soc., 157, B1529 (2010).
34. N. M. Markovic, B. N. Grgur, P. N. Ross, J. Phys. Chem. B, 101, 5404 (1997).
35. B. E. Conway, B. V. Tilak, Electrochim. Acta, 47, 3571 (2002).
36. M. S. Rau, M. R. Gennero de Chialvo, A. C. Chialvo, J. Power Sources, 229, 210
(2013).
37. P. Mandin, T. Pauporte, P. Fanouillere, D. Lincot, J. Electroanal. Chem., 565, 159
(2004).
38. L. Ye, D. Luo, W. Yang, W. Guo, Q. Xu, C. Jiang, J. Hydrog. Energy, 38, 13596
(2013).
39. B. M. Habersberger, T. P. Lodge, F. S. Bates, Macromolecules, 45, 7778 (2012).
40. W. B. Zhang, I. J. Burgess, J. Electroanal. Chem., 668, 66 (2012).
12
Figure Captions
Figure 1
Schematic illustration of experimental setup for measuring the deuterium separation by
using PEFC.
1. H2 gas, 2. D2 gas, 3. O2 gas, 4. Mass flow controller; 5. Gas mixture, 6. Electrolyte
membrane assembly, 7.Variable resistor; 8; Q-mass.
Figure 2
Polarization curves for (a) HER/HOR in 0.05 M H2SO4 and (b) DER/DOR in 0.05 M
D2SO4 on a platinum disk electrode at several rotation rates, 900 rpm (black line), 1600
rpm (blue line), 2500 rpm (green line), and 3600 rpm (red line). Sweep rate is 0.01 V
s-1.
Figure 3
Tafel plots of the reduction current for the HER (dotted line) in 0.05 M H2SO4 and the
DER (solid line) in 0.05 M D2SO4 at low cathodic overpotentials at 298 K. Rotation rate
is 1600 rpm.
Figure 4
Tafel plots of mass transfer-corrected currents for the HOR (dotted line) in 0.05 M
H2SO4 and the DOR (solid line) in 0.05 M D2SO4 at low anodic overpotentials at 298 K.
Rotation rate is 1600 rpm.
Figure 5
Transient behavior of Q-mass ion current for measuring the mixture hydrogen gases
collecting from (a) Line I and (b) Line II through PEFC operated at 1.2 A. (H2 : 20 mL
min-1, D2 : 2 mL min-1)
Figure 6
Dependency of deuterium separation factor on the output current of PEFC operated at
298 K. (H2 : 20 mL min-1, D2 : 2 mL min-1, O2 : 50 mL min-1)
Fig. 1 Schematic illustration of experimental setup for the deuterium separation factor of PEFC. 1. H2 gas, 2. D2 gas, 3. O2 gas, 4. Mass flow controller, 5. Gas mixture unit, 6. Electrolyte membrane assembly, 7.Variable resistor; 8; Q-mass.
2
1
Line I Line II
3 6
5
7
4
4
4
8 8
log i / mA cm-2
-2 -1 0
Pote
ntia
l, Ε
/ V v
s. SC
E
-1.0
-0.9
-0.8
-3 -3 -2 -1 0
-0.7
-0.6
-0.5
-0.4
1 1
Fig. 2 Polarization curves for (a) HER/HOR in 0.05 M H2SO4 and (b) DER/DOR in 0.05 M D2SO4 on a platinum disk electrode at several rotation rates, 900 rpm (black line), 1600 rpm (blue line), 2500 rpm (green line), and 3600 rpm (red line). Sweep rate is 0.01 V s-1.
(a) (b)
log i / mA cm-2
Fig. 3 Tafel plots of the reduction current for the HER (dotted line) in 0.05 M H2SO4 and the DER (solid line) in 0.05 M D2SO4 at low cathodic overpotentials at 298 K. Rotation rate is 1600 rpm.
log i / mA cm-2
-2 -1 0
Ove
rpot
entia
l, η
/ V
-0.15
-0.10
-0.05
0.00
0.034 V/dec
0.039 V/dec
0.143 V/dec
0.168 V/dec
log idi / (id-i) / mA cm-2
Ove
rpot
entia
l, η
/ V
0.00 -2 -1 0
0.02
0.04
0.06
0.08
Fig. 4 Tafel plots of mass transfer-corrected currents for the HOR (dotted line) in 0.05 M H2SO4 and the DOR (solid line) in 0.05 M D2SO4 at low anodic overpotentials at 298 K. Rotation rate is 1600 rpm.
0.055 V/dec
0.091 V/dec
Time, t / hour
Ion
curr
ent,
log(
i)
10-9
(a)
1
10-8
Fig. 5 Transient behavior of Q-mass spectrums for measuring the mixture hydrogen gases collecting from (a) Line I and (b) Line II through PEFC operated at 1.2 A. (H2 : 20 mL min-1, D2 : 2 mL min-1)
10-7
10-6
10-5
(b)
0 2 3 1 0 2 3 4
m = 2
m = 4
m = 3
m = 2
m = 4
m = 3
Time, t / hour
Fig. 6 Dependency of deuterium separation factor on the output current of PEFC operated at 298 K. (H2 : 20 mL min-1, D2 : 2 mL min-1, O2 : 50 mL min-1)
0 0.2 0.4 0.6 0.8 1.0 1.2 0
1
2
3
4
5
Sepa
ratio
n fa
ctor
, α
Output current, J / A